Automatic Detection of Hypoglycemic Events From the Electronic Health Record Notes of Diabetes Patients: Empirical Study. Jin, Y., Li, F., Vimalananda, V. G., & Yu, H. JMIR Medical Informatics, 7(4):e14340, 2019. ![Automatic Detection of Hypoglycemic Events From the Electronic Health Record Notes of Diabetes Patients: Empirical Study [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Background: Hypoglycemic events are common and potentially dangerous conditions among patients being treated for diabetes. Automatic detection of such events could improve patient care and is valuable in population studies. Electronic health records (EHRs) are valuable resources for the detection of such events. Objective: In this study, we aim to develop a deep-learning–based natural language processing (NLP) system to automatically detect hypoglycemic events from EHR notes. Our model is called the High-Performing System for Automatically Detecting Hypoglycemic Events (HYPE). Methods: Domain experts reviewed 500 EHR notes of diabetes patients to determine whether each sentence contained a hypoglycemic event or not. We used this annotated corpus to train and evaluate HYPE, the high-performance NLP system for hypoglycemia detection. We built and evaluated both a classical machine learning model (ie, support vector machines [SVMs]) and state-of-the-art neural network models. Results: We found that neural network models outperformed the SVM model. The convolutional neural network (CNN) model yielded the highest performance in a 10-fold cross-validation setting: mean precision=0.96 (SD 0.03), mean recall=0.86 (SD 0.03), and mean F1=0.91 (SD 0.03). Conclusions: Despite the challenges posed by small and highly imbalanced data, our CNN-based HYPE system still achieved a high performance for hypoglycemia detection. HYPE can be used for EHR-based hypoglycemia surveillance and population studies in diabetes patients. [JMIR Med Inform 2019;7(4):e14340]
@article{jin_automatic_2019,
title = {Automatic {Detection} of {Hypoglycemic} {Events} {From} the {Electronic} {Health} {Record} {Notes} of {Diabetes} {Patients}: {Empirical} {Study}},
volume = {7},
copyright = {Unless stated otherwise, all articles are open-access distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work (},
shorttitle = {Automatic {Detection} of {Hypoglycemic} {Events} {From} the {Electronic} {Health} {Record} {Notes} of {Diabetes} {Patients}},
url = {https://medinform.jmir.org/2019/4/e14340/},
doi = {10.2196/14340},
abstract = {Background: Hypoglycemic events are common and potentially dangerous conditions among patients being treated for diabetes. Automatic detection of such events could improve patient care and is valuable in population studies. Electronic health records (EHRs) are valuable resources for the detection of such events.
Objective: In this study, we aim to develop a deep-learning–based natural language processing (NLP) system to automatically detect hypoglycemic events from EHR notes. Our model is called the High-Performing System for Automatically Detecting Hypoglycemic Events (HYPE).
Methods: Domain experts reviewed 500 EHR notes of diabetes patients to determine whether each sentence contained a hypoglycemic event or not. We used this annotated corpus to train and evaluate HYPE, the high-performance NLP system for hypoglycemia detection. We built and evaluated both a classical machine learning model (ie, support vector machines [SVMs]) and state-of-the-art neural network models.
Results: We found that neural network models outperformed the SVM model. The convolutional neural network (CNN) model yielded the highest performance in a 10-fold cross-validation setting: mean precision=0.96 (SD 0.03), mean recall=0.86 (SD 0.03), and mean F1=0.91 (SD 0.03).
Conclusions: Despite the challenges posed by small and highly imbalanced data, our CNN-based HYPE system still achieved a high performance for hypoglycemia detection. HYPE can be used for EHR-based hypoglycemia surveillance and population studies in diabetes patients.
[JMIR Med Inform 2019;7(4):e14340]},
language = {en},
number = {4},
urldate = {2019-11-10},
journal = {JMIR Medical Informatics},
author = {Jin, Yonghao and Li, Fei and Vimalananda, Varsha G. and Yu, Hong},
year = {2019},
pmid = {31702562 PMCID: PMC6913754},
keywords = {adverse events, convolutional neural networks, hypoglycemia, natural language processing},
pages = {e14340},
}
Downloads: 0
{"_id":"kNkRY5aLQ2dcTwpR6","bibbaseid":"jin-li-vimalananda-yu-automaticdetectionofhypoglycemiceventsfromtheelectronichealthrecordnotesofdiabetespatientsempiricalstudy-2019","author_short":["Jin, Y.","Li, F.","Vimalananda, V. G.","Yu, H."],"bibdata":{"bibtype":"article","type":"article","title":"Automatic Detection of Hypoglycemic Events From the Electronic Health Record Notes of Diabetes Patients: Empirical Study","volume":"7","copyright":"Unless stated otherwise, all articles are open-access distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work (","shorttitle":"Automatic Detection of Hypoglycemic Events From the Electronic Health Record Notes of Diabetes Patients","url":"https://medinform.jmir.org/2019/4/e14340/","doi":"10.2196/14340","abstract":"Background: Hypoglycemic events are common and potentially dangerous conditions among patients being treated for diabetes. Automatic detection of such events could improve patient care and is valuable in population studies. Electronic health records (EHRs) are valuable resources for the detection of such events. Objective: In this study, we aim to develop a deep-learning–based natural language processing (NLP) system to automatically detect hypoglycemic events from EHR notes. Our model is called the High-Performing System for Automatically Detecting Hypoglycemic Events (HYPE). Methods: Domain experts reviewed 500 EHR notes of diabetes patients to determine whether each sentence contained a hypoglycemic event or not. We used this annotated corpus to train and evaluate HYPE, the high-performance NLP system for hypoglycemia detection. We built and evaluated both a classical machine learning model (ie, support vector machines [SVMs]) and state-of-the-art neural network models. Results: We found that neural network models outperformed the SVM model. The convolutional neural network (CNN) model yielded the highest performance in a 10-fold cross-validation setting: mean precision=0.96 (SD 0.03), mean recall=0.86 (SD 0.03), and mean F1=0.91 (SD 0.03). Conclusions: Despite the challenges posed by small and highly imbalanced data, our CNN-based HYPE system still achieved a high performance for hypoglycemia detection. HYPE can be used for EHR-based hypoglycemia surveillance and population studies in diabetes patients. [JMIR Med Inform 2019;7(4):e14340]","language":"en","number":"4","urldate":"2019-11-10","journal":"JMIR Medical Informatics","author":[{"propositions":[],"lastnames":["Jin"],"firstnames":["Yonghao"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Fei"],"suffixes":[]},{"propositions":[],"lastnames":["Vimalananda"],"firstnames":["Varsha","G."],"suffixes":[]},{"propositions":[],"lastnames":["Yu"],"firstnames":["Hong"],"suffixes":[]}],"year":"2019","pmid":"31702562 PMCID: PMC6913754","keywords":"adverse events, convolutional neural networks, hypoglycemia, natural language processing","pages":"e14340","bibtex":"@article{jin_automatic_2019,\n\ttitle = {Automatic {Detection} of {Hypoglycemic} {Events} {From} the {Electronic} {Health} {Record} {Notes} of {Diabetes} {Patients}: {Empirical} {Study}},\n\tvolume = {7},\n\tcopyright = {Unless stated otherwise, all articles are open-access distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work (},\n\tshorttitle = {Automatic {Detection} of {Hypoglycemic} {Events} {From} the {Electronic} {Health} {Record} {Notes} of {Diabetes} {Patients}},\n\turl = {https://medinform.jmir.org/2019/4/e14340/},\n\tdoi = {10.2196/14340},\n\tabstract = {Background: Hypoglycemic events are common and potentially dangerous conditions among patients being treated for diabetes. Automatic detection of such events could improve patient care and is valuable in population studies. Electronic health records (EHRs) are valuable resources for the detection of such events.\n Objective: In this study, we aim to develop a deep-learning–based natural language processing (NLP) system to automatically detect hypoglycemic events from EHR notes. Our model is called the High-Performing System for Automatically Detecting Hypoglycemic Events (HYPE).\n Methods: Domain experts reviewed 500 EHR notes of diabetes patients to determine whether each sentence contained a hypoglycemic event or not. We used this annotated corpus to train and evaluate HYPE, the high-performance NLP system for hypoglycemia detection. We built and evaluated both a classical machine learning model (ie, support vector machines [SVMs]) and state-of-the-art neural network models.\n Results: We found that neural network models outperformed the SVM model. The convolutional neural network (CNN) model yielded the highest performance in a 10-fold cross-validation setting: mean precision=0.96 (SD 0.03), mean recall=0.86 (SD 0.03), and mean F1=0.91 (SD 0.03).\n Conclusions: Despite the challenges posed by small and highly imbalanced data, our CNN-based HYPE system still achieved a high performance for hypoglycemia detection. HYPE can be used for EHR-based hypoglycemia surveillance and population studies in diabetes patients.\n [JMIR Med Inform 2019;7(4):e14340]},\n\tlanguage = {en},\n\tnumber = {4},\n\turldate = {2019-11-10},\n\tjournal = {JMIR Medical Informatics},\n\tauthor = {Jin, Yonghao and Li, Fei and Vimalananda, Varsha G. and Yu, Hong},\n\tyear = {2019},\n\tpmid = {31702562 PMCID: PMC6913754},\n\tkeywords = {adverse events, convolutional neural networks, hypoglycemia, natural language processing},\n\tpages = {e14340},\n}\n\n","author_short":["Jin, Y.","Li, F.","Vimalananda, V. G.","Yu, H."],"key":"jin_automatic_2019","id":"jin_automatic_2019","bibbaseid":"jin-li-vimalananda-yu-automaticdetectionofhypoglycemiceventsfromtheelectronichealthrecordnotesofdiabetespatientsempiricalstudy-2019","role":"author","urls":{"Paper":"https://medinform.jmir.org/2019/4/e14340/"},"keyword":["adverse events","convolutional neural networks","hypoglycemia","natural language processing"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"http://fenway.cs.uml.edu/papers/pubs-all.bib","dataSources":["TqaA9miSB65nRfS5H"],"keywords":["adverse events","convolutional neural networks","hypoglycemia","natural language processing"],"search_terms":["automatic","detection","hypoglycemic","events","electronic","health","record","notes","diabetes","patients","empirical","study","jin","li","vimalananda","yu"],"title":"Automatic Detection of Hypoglycemic Events From the Electronic Health Record Notes of Diabetes Patients: Empirical Study","year":2019}